- Email: [email protected]
Structural analysis of zeolite NaA synthesized by a cost-effective hydrothermal method using kaolin and its use as water softener
Journal of Colloid and Interface Science 367 (2012) 34–39
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Structural analysis of zeolite NaA synthesized by a cost-effective hydrothermal method using kaolin and its use as water softener A.R. Loiola a, J.C.R.A. Andrade a, J.M. Sasaki b, L.R.D. da Silva a,⇑ a
Laboratório de Físico-Química de Minerais e Catálise, Departamento de Química Orgânica e Inorgânica, Universidade Federal do Ceará, C.P. 6002, Campus do Pici, 60451-970 Fortaleza CE, Brazil b Laboratório de Difração de Raios-X, Departamento de Física, Universidade Federal do Ceará, C.P. 6030, Campus do Pici, 60455-760 Fortaleza CE, Brazil
a r t i c l e
i n f o
Article history: Received 15 July 2010 Accepted 11 November 2010 Available online 14 November 2010 Keywords: Zeolite NaA Metakaolin Rietveld refinement Water softener
a b s t r a c t Zeolite 4A (LTA) has been successfully synthesized by a hydrothermal method, where kaolin was used as silica and alumina source. The synthesized zeolite was characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), laser granulometry, and FTIR spectroscopy. XRD data from the Rietveld refinement method confirmed only one crystallographic phase. Zeolite A morphology was observed by SEM analysis, and it showed well-defined crystals with slightly different sizes but with the same cubic shape. Particle size distribution of the crystals was confirmed by laser granulometry, whereas FTIR spectroscopy revealed significant structural differences between the starting material and the final zeolite product used as water softener. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Zeolites are crystalline microporous aluminosilicates with an open framework structure of three-dimensional tetrahedral units generating a network of pores and cavities having molecular dimensions [1]. They are built by SiO2 and AlO4 tetrahedra (or by other tetrahedra such as PO4, GaO4, etc.), which are linked by oxygen atoms [2]. The chemical composition of zeolites is represented by
Ay=m mþ ½ðSiO2 Þx ðAlO2 Þy zH2 O; where A is a cation with charge m, (x + y) is the number of tetrahedra per crystallographic unit cell, and x/y is the silicon:aluminum ratio. Based on the celebrated Löwenstein rule Al–O–Al linkages are not allowed and y/x P 1 [3]. Many zeolites occur as natural minerals, but it is the synthetic varieties which are among the most widely used sorbents, catalysts, and ion-exchange materials in the world [4]. Zeolite A (LTA type) is a synthetic zeolite with very small pores. Zeolite NaA has a pore diameter of 4 Å, which can be modified either to 5 or 3 Å by ion exchange with aqueous solutions of calcium or potassium salts [5]. Zeolite A is normally synthesized in the Na+ form, Na12Al12Si12O48 6 27H2O, and it has a threedimensional pore structure. Its structure is composed of sodalite cages, which are similar to faujasite ones, but connected through double four-membered rings (D4R) of [SiO4]4 and [AlO4]5. By this ⇑ Corresponding author. Fax: +55 85 33669978. E-mail address: [email protected] (L.R.D. da Silva). 0021-9797/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2010.11.026
connection, three cages are present: D4R, sodalite cage, and a-cage. The pore diameter is defined by an eight-member oxygen ring with diameters between 0.23 and 0.42 nm [6]. Due to its low cost and high thermal stability, A-type zeolite has potential applications in separation processes and shape-selective catalysis [5]. The economic importance of zeolites leads scientists to work hard in characterization in order to understand the structures and formation of zeolites, especially to establish the relationship between structure and properties [7]. In fact, there is a great interest in studying zeolite synthesis processes, whose understanding may provide many insights into the mechanism steps. This can give rise to important achievements such as optimization of zeolite industrial production, development of new synthesis techniques, and production of new zeolites for specific purposes [8]. Some properties such as high ion exchange potential, high surface area distributed throughout pores with several diameters, high thermal stability, and high acidity allow zeolites to be materials with great potential for many important applications. Furthermore, they have shown prominence as adsorbents in gas purification, as ions exchangers in detergents, in petroleum refinement catalysis, and in petrochemistry [1]. The use of zeolites as builders in detergents, for instance, requires a competitive price regarding phosphates and other possible sequestrants [9]. Zeolite synthesis from natural sources of SiO4 and Al2O3 shows many advantages in economically and it has been reported in several works [9–12]. Kaolinite is the most important clay mineral of kaolin, whose chemical formula is Al2O3 2SiO22H2O with a 1:1 uncharged dioctahedral layer structure where each layer consists of single silica tetrahedral sheets and single alumina octahedral
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
sheets. It is conveniently used as starting material for zeolite A synthesis once its Si:Al ratio is near to unity which is ideally suited for that of zeolite A [13,14]. Among the several zeolite synthesis methods reported in the literature [1,15–17], the hydrothermal method is the most used. Hydrothermal zeolite synthesis is a multiphase reaction–crystallization process, commonly involving at least one liquid phase and both amorphous and crystalline solid phases [18]. The term hydrothermal is used in a wide sense and includes zeolite crystallization from aqueous systems that contain the necessary chemical components [19]. Zeolites are metastable phases, and slight changes in the synthesis conditions can bring about a pollution of the desired product by the cocrystallization of other phases with a similar composition but with completely different properties (for instance, zeolite A and zeolite X) [20]. In the present work we specifically target zeolite NaA with high purity, by means of a hydrothermal route which is based on kaolinite as starting material, as well as its application in removing Ca2+ ions from solutions. 2. Experimental 2.1. Zeolite synthesis Kaolin supplied by CAULISA S/A, PB-Brazil, was used as Si and Al source. The methodology used to synthesize zeolite A is based on the works of Costa et al. [9,10] and Lucas et al. [12] with modifications in the crystallization and aging time and in the volume used, in order to obtain only one crystalline phase, with uniform particle size. Besides, controlled conditions and SiO2/Al2O3, Na2O/SiO2, and H2O/Na2O appropriate stoichiometric ratios were used [21]. The precursor gel was prepared by mixing 1 g of MK-900 (kaolin previously calcined at 900 °C for 2 h) with 13.3 ml of NaOH 2.75 mol L1. The synthesis was performed by adding the precursor gel to a Teflon-lined autoclave coupled with a water bath set at 70 °C and stirred for 2 h, followed by a resting at 30 °C for 18 h. The resulting material was washed five times with distilled water, centrifuged, and dried at 60 °C overnight. 2.2. Characterization XRD experiments were performed in a Rigaku (DMAXB) X-ray powder diffractometer by using a Bragg–Brentano geometry. The powder patterns were collected in the continuous mode with scan speed of 0.5 min1 (2h). Cu Ka radiation was used with the tube operated at 40 kV and 25 mA. The sample with the particle size below 74 lm (400 mesh) was selected for the diffraction measurement. The Rietveld method [22] was applied using DBWS software for all powder patterns to confirm the structures and also to obtain the full width at half-maximum (FWHM) and to calculate the particle sizes by Scherrer’s equation (Eq. (1)) [23]. FWHM of all peaks, asymmetric coefficients, scale factor, lattice parameters, and background polynomial parameters were refined [24],
L ¼ kk=b cos h;
ð1Þ
where k is the shape coefficient of the reciprocal lattice point (k 1), b (in radians) is the FWHM of the peak, and h is the Bragg angle. The b value, considering a Gaussian distribution for all peaks, was corrected for the instrumental broadening by using the equation
b ¼ ðb2exp b2inst Þ1=2 ;
35
National Institute of Standard Technology) powder standard using the equation [25]
binst ¼ ðU tan2 h þ V tan h þ WÞ1=2 ;
ð3Þ
where U = 0.02806, V = 0.04376, and W = 0.02692 were obtained from output file extracted from Rietveld refinement parameters. Chemical analysis was carried out in a Model SpectraAA 110 Varian spectrometer, by the ICP-AES technique. Kaolinite was dissolved with HCl/NaOH for the determination of Si and Al contents and with HF for other elements such as Fe, Ca, Mg, and Ti. Na and K contents were determined by means of photometry, by using a Model 910 equipment analyzer. Scanning electron microscopy (SEM) analyses were carried out in a Model LX-30 Philips equipment, operated at 12 kV. Zeolite A granulometric distribution was carried out in granulometer CILAS, Model 920 liquid. IR measurements were performed with a Shimadzu IRealise FTIR spectrometer. Samples were prepared as wafers with KBr. The textural characterization of zeolite was done after a 72-h treatment at a temperature of 673 K, by applying N2 adsorption/ desorption isotherms at a temperature of 77 K in a Model ASAP 2020 Micromeritics equipment. From the N2 isotherms acquired, the surface area can be calculated according to the BET and Langmuir model. The final synthesis product obtained from the Paraíba kaolin subjected to an ionic exchange process with the ammonium ion [26] at the back and adding a NH4NO3 10% 10-ml solution into half gram of NaA zeolite and also a 8 h magnetic shake, and then remained untouched for 16 h. Right after the material’s decantation the floating solution remaining was taken out. This described procedure has been repeated for five times. After an ionic exchange with the NH4NO3, a solid sample digestion was made in a TE-0363 TECNAL Kjeldahl microdestilator. Therefore titration with the 5 103 M sulfuric acid solution was done in order to determine the total nitrogen in the sample [27,28]. 2.3. Water softening Softening processes were carried out by using the synthesized zeolite NaA and a commercial zeolite A, donated by Oxanyl Raos Produtos Químicos from São Paulo-Brazil. The NaA zeolite capacity as cation exchanger, removing Ca2+ and delivering Na+, was evaluated taking into account the following parameters: Ca2+ concentration, time of contact between the zeolite and deanalyzed solution, pH, and the relation zeolite weight/Ca2+ concentration. Calcium solutions were prepared by using CaCl22H2O so that solutions with concentrations in a range equivalent to hard water–very hard water have been obtained. Analyses of flame photometry were carried out in order to determine calcium concentrations in the solutions before and after treatment with zeolite NaA, both the synthesized and the commercial one. A910 flame photometer analyzer was used. 2.3.1. Calcium concentration effect Samples of 20 ml containing 80, 90, 100, and 110 ppm of Ca2+ were added in 125 ml Erlenmeyers separately. To each of them was added 50 mg of the synthesized zeolite NaA. The recipients were mechanically stirred for 2 h and after 30 min, the solutions were filtered and their concentrations determined by flame photometry. The same procedure was made for the commercial zeolite. All the experiments were carried out in triplicate.
ð2Þ
where bexp is the measured broadening and binst is the broadening due to the instrument. binst was obtained from LaB6 (SRM660–
2.3.2. Effect of time Samples of 20 ml of 80 ppm Ca2+ solution were added to 125 ml Erlenmeyers. To each recipient was added 50 mg of synthesized
36
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
zeolite. The samples were stirred during different times, from 1 min until 24 h. The solutions were filtered and analyzed by flame photometry. This procedure was repeated using the commercial zeolite.
2.3.3. Effect of pH The amount of 80 ppm Ca2+ solutions had their pH varying from 2 to 10, adjusted by using HCl and NaOH solution. Samples of 20 ml were transferred to 125 ml Erlenmeyers, in which 50 mg of the synthesized zeolite was added. The samples were stirred for 2 h and filtered. Commercial zeolite was also used.
2.3.4. Effect of zeolite weight Samples of 20 ml of Ca2+ solution were transferred to 125 ml Erlenmeyers in which amounts of 5, 10, 20, 50, 80, and 100 mg of the synthesized zeolite were added. The same amount of commercial zeolite was used in similar experiments. The samples were stirred for 2 h, filtered, and analyzed by means of flame photometry.
Fig. 1. (a) Experimental XRD pattern of kaolinite, (b) calculated XRD pattern of kaolinite by Rietveld refinement, and (c) the difference between the experimental and calculated patterns.
3. Results and discussion Chemical analysis for kaolin showed the following chemical formula: (Fe0.34Mg0.03Na0.03K0.03)Al4Si4O10(OH)8. Kaolin needs to be thermally activated in order to become reactive within the reaction mixture. This is promptly achieved by means of calcination, which provides enough energy to promote the required changes in the thermodynamic conditions. This process of activation results in drastic changes in the kaolin structure, driving it to lose its crystallinity. In the kaolin structure the Al atoms are octahedrally coordinated to two vertex oxygens from the SiO4 tetrahedral layer, to an (OH) group from one side and three (OH) groups from the parallel hydroxyl layer from the other side [29]. In the metakaolin, on the other hand, the Al atoms lose their initial configuration and are tetrahedrically coordinated. This scenario is closer to the zeolite structure than that observed in the kaolin structure; thus metakaolin is suitable as raw material for zeolite synthesis.
3.1. Structural refinement XRD data refinement is an important tool which can be used to evaluate the structural aspects of a crystalline material. It is especially useful for verifying the number of crystalline phases present in a sample. The raw kaolin XRD measured pattern is shown in Fig. 1a. The calculated XRD pattern is presented in Fig. 1b and the difference between measured and calculated patterns in Fig. 1c. These results revealed the presence of only the clay mineral kaolinite without other crystalline phases, which suggests an appropriate composition as raw material to be used in the synthesis of zeolite. Fig. 2 shows the experimental, simulated, and difference powder patterns of the synthesized zeolite A. From the refinement it is clear that zeolite A belongs to the cubic crystalline system and the space group Fm3c [30] with the lattice parameters: a = b = c = 24.61 Å and a = b = c = 90°. The final results of the Rietveld analysis are the following: the value of the expected error (R) is 4.05% and the obtained error (R WP) is 13.45%. The R WP/R ratio (S) is 3.31, which means a high degree of certainty for the identified phase. Only one crystalline phase of zeolite A was achieved. From an economic point of view it is a major advantage as it is a fast process which requires cheap substances, and which opens even more space for applications which require it in large quantities.
Fig. 2. (a) Experimental XRD pattern of the synthesized zeolite NaA, (b) calculated XRD pattern of synthesized zeolite NaA by Rietveld refinement, and (c) the difference between the experimental and the calculated patterns.
3.2. Scanning electron microscopy Scanning electron micrographs of the synthesized zeolite (Fig. 3) revealed the presence of crystals with the same cubic morphology and with slightly different sizes. Fig. 3b shows a highmagnification SEM image. The morphology of this zeolite crystal is cubic, with an average diameter of approximately 2 lm. Additionally, no considerable amount of amorphous materials was detected by this technique, which indicates a high crystallization degree and that the methodology described in this work is advantageous as a cost-effective alternative to produce zeolite A. 3.3. Particle size distribution The results of the granulometric distribution analysis are shown in Fig. 4. The curve of the granulometric distribution (sigmoidal line) indicates the sum of the particles, in percentage, referring to each particle size. The analysis of this curve leads to two
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
37
Fig. 5. FTIR of (a) kaolin, (b) metakaolin, and (c) zeolite NaA.
Fig. 3. SEM of zeolite NaA with different magnifications.
representative points: the first one at approximately 3 lm and the second one around 12 lm. These values indicate the average sizes of the particles. The first value is in agreement with the results obtained from scanning electron microscopy; the second value, on the other hand, differs from what the SEM results indicates, but can be associated to agglomerates formed by zeolite crystals. 3.4. FTIR The transformation of kaolin to metakaolin and then to zeolite A can be clearly observed from IR spectra (Fig. 5) in the lattice region (1400–400 cm1). The kaolin starting material gives at least 20 well-defined IR bands in this region due to Si–O, Si–O–Al, and Al–OH vibrations. The conversion to metakaolin totally removes these bands, leaving a broad intense asymmetric band at 1095 cm1 as the major feature. The disappearance of the 914 and 937 cm1 bands indicates the loss of Al–OH units, while the
changes in Si–O stretching bands and the disappearance of the Si–O–Al bands at 793 and 756 cm1 are consistent with distortion of the tetrahedral and octahedral layers [31,32]. In the FTIR spectrum of the zeolite NaA, the characteristic bands for zeolite framework at 557 cm1 due to the external vibration of double four-rings, 1001 cm1 for the internal vibration of (Si, Al)–O asymmetric stretching, 671 cm1 for the internal vibration of (Si, Al)–O symmetric stretching, and 467 cm1 for the internal vibration of (Si, Al)–O bending were observed. The band related to OH also appeared at about 1655 cm1 [33–36]. 3.5. Surface area From N2 adsorption/desorption isotherms, the surface area can be determined by using the BET (9.20 m2/g) and Langmuir (10.32 m2/g) method. Whenever a cationic exchange happens, the surface area increases significantly according to the exchanged cation [38]. 3.6. Cationic exchange capacity (CEC) The nitrogen distillation procedure using the Kjeldahl method is made by the sample digestion in order to change all the nitrogen into ammonium ion, applying MgO.
Fig. 4. Zeolite NaA granulometric distribution.
38
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
3.7. NaOH
mine the ideal zeolite amount which should be used in order to obtain the best yield. According to this graphic, better perfor-
In order to alkalinize the environment, NaOH, 40% m/v, is added during the procedure. Thus ammonia is produced and after that it is distilled by a dragging water steam. Then it is mixed in a 2% m/v boric acid solution with the bromocresol and red methyl indicators to spot the final point in the retrotitration of the formed borate ion, with a 0.1 mol L1 standardized cloridric acid solution. The quantity of ammonia found in the sample, the CEC, was 920.0 mg/ 100 g of zeolite. 3.8. Water softening 3.8.1. Calcium concentration effect Although both zeolites presented good results as water softeners, the synthesized zeolite was more efficient in removing Ca2+, lowering the hardness practically to zero, as seen in Fig. 6. The Ca2+ concentration does not seem to exert influence over the synthesized zeolite efficiency, different from that observed in the commercial zeolite where the efficiency decreases as the Ca2+ concentration is increased. 3.8.2. Effect of time The effect of time over the efficiency of zeolite NaA as water softener is shown in Fig. 7. It is observed that even with 1 min of treatment, the synthesized zeolite reduced the Ca2+ concentration from 80 to 3 ppm, equivalent to a reduction of ca. 96%. On the other hand, the commercial zeolite gives rise to a hardness reduction of only 75% during the same time, making it less efficient for this purpose. From 1 h on, the hardness is completely reduced when the synthesized zeolite is used. Residues of Ca2+ (ca. 5% of the initial concentration) are observed when the solution is treated with commercial zeolite.
Fig. 7. Effect of time: h, pattern solution; s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
3.8.3. Effect of pH The effect of pH over the efficiency of zeolite NaA in the process of water softening is shown in Fig. 8. No considerable effect is observed, which indicates a high efficiency of the zeolite in a broad range of pH values. Again, better results are observed for the synthesized zeolite. 3.8.4. Effect of weight Fig. 9 shows the softening results obtained by varying the weight of zeolite. The analysis of these data is essential to deter-
Fig. 8. Effect of pH: h, pattern solution; s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
Fig. 6. Effect of Ca2+ concentration: h, pattern solution; s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
Fig. 9. Effect of zeolite weight: s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
mances of the zeolite are achieved as its weight is increased. However, from 20 mg on, no appreciable improvement is observed. For the commercial zeolite higher quantities are necessary to achieve equivalent levels of reduction in the water hardness. Under these circumstances, the capacity of the synthesized zeolite toward calcium removal reached 31 mg calcium/g zeolite, which is almost twice that observed for the commercial zeolite (17 mg calcium/g zeolite), and three times more efficient than a similar natural zeolite [39]. 4. Conclusions The high intensity of the peaks observed in the XRD powder patterns indicates that the synthesized zeolite presents high crystallinity. XRD data refinement by the Rietveld method confirmed that only one phase was obtained by the hydrothermal process. SEM analyses clearly showed the single crystals of zeolite NaA. High resolution (20,000) SEM images of zeolite A show the cubic morphology of the material, which agrees with current literature data. Finally, the transformation from kaolin to zeolite NaA could be monitored and evaluated by means of FTIR. All these results show that the zeolite synthesis method, based on the use of kaolin as raw material and on fast processes, is highly efficient. The results from the softening analysis clearly show better yields for the synthesized zeolite than for the commercial zeolite. A reduction of 95% in the water hardness in 1 L of solution requires 1 g of the synthesized zeolite, while 4 g of commercial zeolite is necessary for similar reduction in the same solution. Besides, smaller times are required for the synthesized zeolite to achieve maximum performance. Acknowledgments The authors thank Dr. Pablo Cubillas for spelling corrections and the Brazilian Ministry of Education-CAPES for the financial support.
39
References [1] D.W. Breck, Zeolites: Molecular Sieves, first ed., Wiley, New York, 1974. [2] M. Guisnet, F.R. Ribeiro, Zeólitos: Um Nanomundo ao Serviço da Catálise, first ed., Fundação Calouste Gulbenkian, Lisboa, 2004. [3] K. Shams, S.J. Mirmohammadi, Microporous Mesoporous Mater. 106 (2007) 268–277. [4] C.S. Cundy, Chem. Rev. 103 (2003) 663–701. [5] H. Yang et al., Microporous Mesoporous Mater. 117 (2009) 33–40. [6] T.M. Salama et al., Mater. Chem. Phys. 113 (2009) 159–165. [7] D.P. Serrano et al., Microporous Mesoporous Mater. 69 (2004) 197–208. [8] M. Schmachtl et al., Ultrasonics 38 (2000) 809–812. [9] E. Costa et al., Ind. Eng. Chem. Res. 27 (1988) 1291–1296. [10] E. Costa et al., An. Quim. 84 (1987) 366–373. [11] R.M. Barrer, D.M. MacLeod, Faraday Soc. 51 (1955) 1290–1300. [12] A. Lucas et al., Ind. Eng. Chem. Res. 31 (1992) 2134–2140. [13] D.-C. Lin et al., Microporous Mesoporous Mater. 70 (2004) 63–70. [14] H. Youssef, D. Ibrahim, S. Komarneni, Microporous Mesoporous Mater. 115 (2009) 527–534. [15] S. Chandrasekhar, P.N. Pramada, Porous Mater. 6 (1999) 283–297. [16] J. Aguado et al., Microporous Mesoporous Mater. 75 (2004) 41–49. [17] D. Boukadir, N. Bettahar, Z. Derriche, Ann. Chim. Sci. Mater. 227 (2002) 1–13. [18] C.S. Cundy, P.A. Cox, Microporous Mesoporous Mater. 82 (2005) 1–78. [19] Y. Hu, X. Liu, Miner. Eng. 16 (2003) 1279–1284. [20] C. Bebon et al., Microporous Mesoporous Mater. 53 (2002) 13–20. [21] S. Chandrasekhar, P.N. Pramada, Microporous Mesoporous Mater. 108 (2008) 152–161. [22] H.M. Rietveld, Acta Crystallogr. 22 (1967) 151–152. [23] R.A. Young, J. Appl. Crystallogr. 28 (1995) 366–367. [24] L.V. Azároff, Elements of X-Ray Crystallography, first ed., McGraw-Hill, New York, 1968. [25] M.J. CagliotI, A. Paoletti, F.P. Ricci, Nucl. Instrum. Methods 35 (1958) 223–228. [26] S. Chandrasekhar, P.N. Pramada, Ceram. Int. 27 (2001) 351–361. [27] L.I.D. da Silva et al., Quím. Nova 29 (2006) 46–51. [28] S.A. Khan, R.L. Mulvaney, R.G. Hoeft, Soc. Am. J. 64 (2000) 1083–1089. [29] R. Dimitrijevic, W. Lutz, A. Ritzmann, J. Phys. Chem. 67 (2006) 1741–1748. [30] V. Gramlich, Z. Meier, Kristallogr., Kristallgeom., Kristallphys., Kristallchem. 133 (1971) 134–149. [31] C. Breen et al., Vib. Spectrosc. 43 (2007) 366–379. [32] D. Akolkar, A. Chaffee, R.F. Howe, Zeolites 19 (1997) 359–365. [33] R.W. Stevens, R.V. Siriwardane, J. Logan, Energy Fuels 22 (2008) 3070–3079. [34] H. Yamada et al., Sci. Technol. Adv. Mater. 6 (2005) 394–398. [35] Y. Huang, Z. Jiang, Microporous Mater. 12 (1997) 341–345. [36] K.A. Iyer, S.J. Singer, Phys. Chem. 98 (1994) 12679–12686. [38] A.A. Ismail et al., Colloids Surf. A 366 (2010) 80–87. [39] S. Cinar, B. Beler-Baykal, Water Sci. Technol. 51 (2005) 71–77.
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Structural analysis of zeolite NaA synthesized by a cost-effective hydrothermal method using kaolin and its use as water softener A.R. Loiola a, J.C.R.A. Andrade a, J.M. Sasaki b, L.R.D. da Silva a,⇑ a
Laboratório de Físico-Química de Minerais e Catálise, Departamento de Química Orgânica e Inorgânica, Universidade Federal do Ceará, C.P. 6002, Campus do Pici, 60451-970 Fortaleza CE, Brazil b Laboratório de Difração de Raios-X, Departamento de Física, Universidade Federal do Ceará, C.P. 6030, Campus do Pici, 60455-760 Fortaleza CE, Brazil
a r t i c l e
i n f o
Article history: Received 15 July 2010 Accepted 11 November 2010 Available online 14 November 2010 Keywords: Zeolite NaA Metakaolin Rietveld refinement Water softener
a b s t r a c t Zeolite 4A (LTA) has been successfully synthesized by a hydrothermal method, where kaolin was used as silica and alumina source. The synthesized zeolite was characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), laser granulometry, and FTIR spectroscopy. XRD data from the Rietveld refinement method confirmed only one crystallographic phase. Zeolite A morphology was observed by SEM analysis, and it showed well-defined crystals with slightly different sizes but with the same cubic shape. Particle size distribution of the crystals was confirmed by laser granulometry, whereas FTIR spectroscopy revealed significant structural differences between the starting material and the final zeolite product used as water softener. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Zeolites are crystalline microporous aluminosilicates with an open framework structure of three-dimensional tetrahedral units generating a network of pores and cavities having molecular dimensions [1]. They are built by SiO2 and AlO4 tetrahedra (or by other tetrahedra such as PO4, GaO4, etc.), which are linked by oxygen atoms [2]. The chemical composition of zeolites is represented by
Ay=m mþ ½ðSiO2 Þx ðAlO2 Þy zH2 O; where A is a cation with charge m, (x + y) is the number of tetrahedra per crystallographic unit cell, and x/y is the silicon:aluminum ratio. Based on the celebrated Löwenstein rule Al–O–Al linkages are not allowed and y/x P 1 [3]. Many zeolites occur as natural minerals, but it is the synthetic varieties which are among the most widely used sorbents, catalysts, and ion-exchange materials in the world [4]. Zeolite A (LTA type) is a synthetic zeolite with very small pores. Zeolite NaA has a pore diameter of 4 Å, which can be modified either to 5 or 3 Å by ion exchange with aqueous solutions of calcium or potassium salts [5]. Zeolite A is normally synthesized in the Na+ form, Na12Al12Si12O48 6 27H2O, and it has a threedimensional pore structure. Its structure is composed of sodalite cages, which are similar to faujasite ones, but connected through double four-membered rings (D4R) of [SiO4]4 and [AlO4]5. By this ⇑ Corresponding author. Fax: +55 85 33669978. E-mail address: [email protected] (L.R.D. da Silva). 0021-9797/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2010.11.026
connection, three cages are present: D4R, sodalite cage, and a-cage. The pore diameter is defined by an eight-member oxygen ring with diameters between 0.23 and 0.42 nm [6]. Due to its low cost and high thermal stability, A-type zeolite has potential applications in separation processes and shape-selective catalysis [5]. The economic importance of zeolites leads scientists to work hard in characterization in order to understand the structures and formation of zeolites, especially to establish the relationship between structure and properties [7]. In fact, there is a great interest in studying zeolite synthesis processes, whose understanding may provide many insights into the mechanism steps. This can give rise to important achievements such as optimization of zeolite industrial production, development of new synthesis techniques, and production of new zeolites for specific purposes [8]. Some properties such as high ion exchange potential, high surface area distributed throughout pores with several diameters, high thermal stability, and high acidity allow zeolites to be materials with great potential for many important applications. Furthermore, they have shown prominence as adsorbents in gas purification, as ions exchangers in detergents, in petroleum refinement catalysis, and in petrochemistry [1]. The use of zeolites as builders in detergents, for instance, requires a competitive price regarding phosphates and other possible sequestrants [9]. Zeolite synthesis from natural sources of SiO4 and Al2O3 shows many advantages in economically and it has been reported in several works [9–12]. Kaolinite is the most important clay mineral of kaolin, whose chemical formula is Al2O3 2SiO22H2O with a 1:1 uncharged dioctahedral layer structure where each layer consists of single silica tetrahedral sheets and single alumina octahedral
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
sheets. It is conveniently used as starting material for zeolite A synthesis once its Si:Al ratio is near to unity which is ideally suited for that of zeolite A [13,14]. Among the several zeolite synthesis methods reported in the literature [1,15–17], the hydrothermal method is the most used. Hydrothermal zeolite synthesis is a multiphase reaction–crystallization process, commonly involving at least one liquid phase and both amorphous and crystalline solid phases [18]. The term hydrothermal is used in a wide sense and includes zeolite crystallization from aqueous systems that contain the necessary chemical components [19]. Zeolites are metastable phases, and slight changes in the synthesis conditions can bring about a pollution of the desired product by the cocrystallization of other phases with a similar composition but with completely different properties (for instance, zeolite A and zeolite X) [20]. In the present work we specifically target zeolite NaA with high purity, by means of a hydrothermal route which is based on kaolinite as starting material, as well as its application in removing Ca2+ ions from solutions. 2. Experimental 2.1. Zeolite synthesis Kaolin supplied by CAULISA S/A, PB-Brazil, was used as Si and Al source. The methodology used to synthesize zeolite A is based on the works of Costa et al. [9,10] and Lucas et al. [12] with modifications in the crystallization and aging time and in the volume used, in order to obtain only one crystalline phase, with uniform particle size. Besides, controlled conditions and SiO2/Al2O3, Na2O/SiO2, and H2O/Na2O appropriate stoichiometric ratios were used [21]. The precursor gel was prepared by mixing 1 g of MK-900 (kaolin previously calcined at 900 °C for 2 h) with 13.3 ml of NaOH 2.75 mol L1. The synthesis was performed by adding the precursor gel to a Teflon-lined autoclave coupled with a water bath set at 70 °C and stirred for 2 h, followed by a resting at 30 °C for 18 h. The resulting material was washed five times with distilled water, centrifuged, and dried at 60 °C overnight. 2.2. Characterization XRD experiments were performed in a Rigaku (DMAXB) X-ray powder diffractometer by using a Bragg–Brentano geometry. The powder patterns were collected in the continuous mode with scan speed of 0.5 min1 (2h). Cu Ka radiation was used with the tube operated at 40 kV and 25 mA. The sample with the particle size below 74 lm (400 mesh) was selected for the diffraction measurement. The Rietveld method [22] was applied using DBWS software for all powder patterns to confirm the structures and also to obtain the full width at half-maximum (FWHM) and to calculate the particle sizes by Scherrer’s equation (Eq. (1)) [23]. FWHM of all peaks, asymmetric coefficients, scale factor, lattice parameters, and background polynomial parameters were refined [24],
L ¼ kk=b cos h;
ð1Þ
where k is the shape coefficient of the reciprocal lattice point (k 1), b (in radians) is the FWHM of the peak, and h is the Bragg angle. The b value, considering a Gaussian distribution for all peaks, was corrected for the instrumental broadening by using the equation
b ¼ ðb2exp b2inst Þ1=2 ;
35
National Institute of Standard Technology) powder standard using the equation [25]
binst ¼ ðU tan2 h þ V tan h þ WÞ1=2 ;
ð3Þ
where U = 0.02806, V = 0.04376, and W = 0.02692 were obtained from output file extracted from Rietveld refinement parameters. Chemical analysis was carried out in a Model SpectraAA 110 Varian spectrometer, by the ICP-AES technique. Kaolinite was dissolved with HCl/NaOH for the determination of Si and Al contents and with HF for other elements such as Fe, Ca, Mg, and Ti. Na and K contents were determined by means of photometry, by using a Model 910 equipment analyzer. Scanning electron microscopy (SEM) analyses were carried out in a Model LX-30 Philips equipment, operated at 12 kV. Zeolite A granulometric distribution was carried out in granulometer CILAS, Model 920 liquid. IR measurements were performed with a Shimadzu IRealise FTIR spectrometer. Samples were prepared as wafers with KBr. The textural characterization of zeolite was done after a 72-h treatment at a temperature of 673 K, by applying N2 adsorption/ desorption isotherms at a temperature of 77 K in a Model ASAP 2020 Micromeritics equipment. From the N2 isotherms acquired, the surface area can be calculated according to the BET and Langmuir model. The final synthesis product obtained from the Paraíba kaolin subjected to an ionic exchange process with the ammonium ion [26] at the back and adding a NH4NO3 10% 10-ml solution into half gram of NaA zeolite and also a 8 h magnetic shake, and then remained untouched for 16 h. Right after the material’s decantation the floating solution remaining was taken out. This described procedure has been repeated for five times. After an ionic exchange with the NH4NO3, a solid sample digestion was made in a TE-0363 TECNAL Kjeldahl microdestilator. Therefore titration with the 5 103 M sulfuric acid solution was done in order to determine the total nitrogen in the sample [27,28]. 2.3. Water softening Softening processes were carried out by using the synthesized zeolite NaA and a commercial zeolite A, donated by Oxanyl Raos Produtos Químicos from São Paulo-Brazil. The NaA zeolite capacity as cation exchanger, removing Ca2+ and delivering Na+, was evaluated taking into account the following parameters: Ca2+ concentration, time of contact between the zeolite and deanalyzed solution, pH, and the relation zeolite weight/Ca2+ concentration. Calcium solutions were prepared by using CaCl22H2O so that solutions with concentrations in a range equivalent to hard water–very hard water have been obtained. Analyses of flame photometry were carried out in order to determine calcium concentrations in the solutions before and after treatment with zeolite NaA, both the synthesized and the commercial one. A910 flame photometer analyzer was used. 2.3.1. Calcium concentration effect Samples of 20 ml containing 80, 90, 100, and 110 ppm of Ca2+ were added in 125 ml Erlenmeyers separately. To each of them was added 50 mg of the synthesized zeolite NaA. The recipients were mechanically stirred for 2 h and after 30 min, the solutions were filtered and their concentrations determined by flame photometry. The same procedure was made for the commercial zeolite. All the experiments were carried out in triplicate.
ð2Þ
where bexp is the measured broadening and binst is the broadening due to the instrument. binst was obtained from LaB6 (SRM660–
2.3.2. Effect of time Samples of 20 ml of 80 ppm Ca2+ solution were added to 125 ml Erlenmeyers. To each recipient was added 50 mg of synthesized
36
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
zeolite. The samples were stirred during different times, from 1 min until 24 h. The solutions were filtered and analyzed by flame photometry. This procedure was repeated using the commercial zeolite.
2.3.3. Effect of pH The amount of 80 ppm Ca2+ solutions had their pH varying from 2 to 10, adjusted by using HCl and NaOH solution. Samples of 20 ml were transferred to 125 ml Erlenmeyers, in which 50 mg of the synthesized zeolite was added. The samples were stirred for 2 h and filtered. Commercial zeolite was also used.
2.3.4. Effect of zeolite weight Samples of 20 ml of Ca2+ solution were transferred to 125 ml Erlenmeyers in which amounts of 5, 10, 20, 50, 80, and 100 mg of the synthesized zeolite were added. The same amount of commercial zeolite was used in similar experiments. The samples were stirred for 2 h, filtered, and analyzed by means of flame photometry.
Fig. 1. (a) Experimental XRD pattern of kaolinite, (b) calculated XRD pattern of kaolinite by Rietveld refinement, and (c) the difference between the experimental and calculated patterns.
3. Results and discussion Chemical analysis for kaolin showed the following chemical formula: (Fe0.34Mg0.03Na0.03K0.03)Al4Si4O10(OH)8. Kaolin needs to be thermally activated in order to become reactive within the reaction mixture. This is promptly achieved by means of calcination, which provides enough energy to promote the required changes in the thermodynamic conditions. This process of activation results in drastic changes in the kaolin structure, driving it to lose its crystallinity. In the kaolin structure the Al atoms are octahedrally coordinated to two vertex oxygens from the SiO4 tetrahedral layer, to an (OH) group from one side and three (OH) groups from the parallel hydroxyl layer from the other side [29]. In the metakaolin, on the other hand, the Al atoms lose their initial configuration and are tetrahedrically coordinated. This scenario is closer to the zeolite structure than that observed in the kaolin structure; thus metakaolin is suitable as raw material for zeolite synthesis.
3.1. Structural refinement XRD data refinement is an important tool which can be used to evaluate the structural aspects of a crystalline material. It is especially useful for verifying the number of crystalline phases present in a sample. The raw kaolin XRD measured pattern is shown in Fig. 1a. The calculated XRD pattern is presented in Fig. 1b and the difference between measured and calculated patterns in Fig. 1c. These results revealed the presence of only the clay mineral kaolinite without other crystalline phases, which suggests an appropriate composition as raw material to be used in the synthesis of zeolite. Fig. 2 shows the experimental, simulated, and difference powder patterns of the synthesized zeolite A. From the refinement it is clear that zeolite A belongs to the cubic crystalline system and the space group Fm3c [30] with the lattice parameters: a = b = c = 24.61 Å and a = b = c = 90°. The final results of the Rietveld analysis are the following: the value of the expected error (R) is 4.05% and the obtained error (R WP) is 13.45%. The R WP/R ratio (S) is 3.31, which means a high degree of certainty for the identified phase. Only one crystalline phase of zeolite A was achieved. From an economic point of view it is a major advantage as it is a fast process which requires cheap substances, and which opens even more space for applications which require it in large quantities.
Fig. 2. (a) Experimental XRD pattern of the synthesized zeolite NaA, (b) calculated XRD pattern of synthesized zeolite NaA by Rietveld refinement, and (c) the difference between the experimental and the calculated patterns.
3.2. Scanning electron microscopy Scanning electron micrographs of the synthesized zeolite (Fig. 3) revealed the presence of crystals with the same cubic morphology and with slightly different sizes. Fig. 3b shows a highmagnification SEM image. The morphology of this zeolite crystal is cubic, with an average diameter of approximately 2 lm. Additionally, no considerable amount of amorphous materials was detected by this technique, which indicates a high crystallization degree and that the methodology described in this work is advantageous as a cost-effective alternative to produce zeolite A. 3.3. Particle size distribution The results of the granulometric distribution analysis are shown in Fig. 4. The curve of the granulometric distribution (sigmoidal line) indicates the sum of the particles, in percentage, referring to each particle size. The analysis of this curve leads to two
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
37
Fig. 5. FTIR of (a) kaolin, (b) metakaolin, and (c) zeolite NaA.
Fig. 3. SEM of zeolite NaA with different magnifications.
representative points: the first one at approximately 3 lm and the second one around 12 lm. These values indicate the average sizes of the particles. The first value is in agreement with the results obtained from scanning electron microscopy; the second value, on the other hand, differs from what the SEM results indicates, but can be associated to agglomerates formed by zeolite crystals. 3.4. FTIR The transformation of kaolin to metakaolin and then to zeolite A can be clearly observed from IR spectra (Fig. 5) in the lattice region (1400–400 cm1). The kaolin starting material gives at least 20 well-defined IR bands in this region due to Si–O, Si–O–Al, and Al–OH vibrations. The conversion to metakaolin totally removes these bands, leaving a broad intense asymmetric band at 1095 cm1 as the major feature. The disappearance of the 914 and 937 cm1 bands indicates the loss of Al–OH units, while the
changes in Si–O stretching bands and the disappearance of the Si–O–Al bands at 793 and 756 cm1 are consistent with distortion of the tetrahedral and octahedral layers [31,32]. In the FTIR spectrum of the zeolite NaA, the characteristic bands for zeolite framework at 557 cm1 due to the external vibration of double four-rings, 1001 cm1 for the internal vibration of (Si, Al)–O asymmetric stretching, 671 cm1 for the internal vibration of (Si, Al)–O symmetric stretching, and 467 cm1 for the internal vibration of (Si, Al)–O bending were observed. The band related to OH also appeared at about 1655 cm1 [33–36]. 3.5. Surface area From N2 adsorption/desorption isotherms, the surface area can be determined by using the BET (9.20 m2/g) and Langmuir (10.32 m2/g) method. Whenever a cationic exchange happens, the surface area increases significantly according to the exchanged cation [38]. 3.6. Cationic exchange capacity (CEC) The nitrogen distillation procedure using the Kjeldahl method is made by the sample digestion in order to change all the nitrogen into ammonium ion, applying MgO.
Fig. 4. Zeolite NaA granulometric distribution.
38
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
3.7. NaOH
mine the ideal zeolite amount which should be used in order to obtain the best yield. According to this graphic, better perfor-
In order to alkalinize the environment, NaOH, 40% m/v, is added during the procedure. Thus ammonia is produced and after that it is distilled by a dragging water steam. Then it is mixed in a 2% m/v boric acid solution with the bromocresol and red methyl indicators to spot the final point in the retrotitration of the formed borate ion, with a 0.1 mol L1 standardized cloridric acid solution. The quantity of ammonia found in the sample, the CEC, was 920.0 mg/ 100 g of zeolite. 3.8. Water softening 3.8.1. Calcium concentration effect Although both zeolites presented good results as water softeners, the synthesized zeolite was more efficient in removing Ca2+, lowering the hardness practically to zero, as seen in Fig. 6. The Ca2+ concentration does not seem to exert influence over the synthesized zeolite efficiency, different from that observed in the commercial zeolite where the efficiency decreases as the Ca2+ concentration is increased. 3.8.2. Effect of time The effect of time over the efficiency of zeolite NaA as water softener is shown in Fig. 7. It is observed that even with 1 min of treatment, the synthesized zeolite reduced the Ca2+ concentration from 80 to 3 ppm, equivalent to a reduction of ca. 96%. On the other hand, the commercial zeolite gives rise to a hardness reduction of only 75% during the same time, making it less efficient for this purpose. From 1 h on, the hardness is completely reduced when the synthesized zeolite is used. Residues of Ca2+ (ca. 5% of the initial concentration) are observed when the solution is treated with commercial zeolite.
Fig. 7. Effect of time: h, pattern solution; s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
3.8.3. Effect of pH The effect of pH over the efficiency of zeolite NaA in the process of water softening is shown in Fig. 8. No considerable effect is observed, which indicates a high efficiency of the zeolite in a broad range of pH values. Again, better results are observed for the synthesized zeolite. 3.8.4. Effect of weight Fig. 9 shows the softening results obtained by varying the weight of zeolite. The analysis of these data is essential to deter-
Fig. 8. Effect of pH: h, pattern solution; s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
Fig. 6. Effect of Ca2+ concentration: h, pattern solution; s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
Fig. 9. Effect of zeolite weight: s, after treatment with synthesized zeolite; 4, after treatment with commercial zeolite.
A.R. Loiola et al. / Journal of Colloid and Interface Science 367 (2012) 34–39
mances of the zeolite are achieved as its weight is increased. However, from 20 mg on, no appreciable improvement is observed. For the commercial zeolite higher quantities are necessary to achieve equivalent levels of reduction in the water hardness. Under these circumstances, the capacity of the synthesized zeolite toward calcium removal reached 31 mg calcium/g zeolite, which is almost twice that observed for the commercial zeolite (17 mg calcium/g zeolite), and three times more efficient than a similar natural zeolite [39]. 4. Conclusions The high intensity of the peaks observed in the XRD powder patterns indicates that the synthesized zeolite presents high crystallinity. XRD data refinement by the Rietveld method confirmed that only one phase was obtained by the hydrothermal process. SEM analyses clearly showed the single crystals of zeolite NaA. High resolution (20,000) SEM images of zeolite A show the cubic morphology of the material, which agrees with current literature data. Finally, the transformation from kaolin to zeolite NaA could be monitored and evaluated by means of FTIR. All these results show that the zeolite synthesis method, based on the use of kaolin as raw material and on fast processes, is highly efficient. The results from the softening analysis clearly show better yields for the synthesized zeolite than for the commercial zeolite. A reduction of 95% in the water hardness in 1 L of solution requires 1 g of the synthesized zeolite, while 4 g of commercial zeolite is necessary for similar reduction in the same solution. Besides, smaller times are required for the synthesized zeolite to achieve maximum performance. Acknowledgments The authors thank Dr. Pablo Cubillas for spelling corrections and the Brazilian Ministry of Education-CAPES for the financial support.
39
References [1] D.W. Breck, Zeolites: Molecular Sieves, first ed., Wiley, New York, 1974. [2] M. Guisnet, F.R. Ribeiro, Zeólitos: Um Nanomundo ao Serviço da Catálise, first ed., Fundação Calouste Gulbenkian, Lisboa, 2004. [3] K. Shams, S.J. Mirmohammadi, Microporous Mesoporous Mater. 106 (2007) 268–277. [4] C.S. Cundy, Chem. Rev. 103 (2003) 663–701. [5] H. Yang et al., Microporous Mesoporous Mater. 117 (2009) 33–40. [6] T.M. Salama et al., Mater. Chem. Phys. 113 (2009) 159–165. [7] D.P. Serrano et al., Microporous Mesoporous Mater. 69 (2004) 197–208. [8] M. Schmachtl et al., Ultrasonics 38 (2000) 809–812. [9] E. Costa et al., Ind. Eng. Chem. Res. 27 (1988) 1291–1296. [10] E. Costa et al., An. Quim. 84 (1987) 366–373. [11] R.M. Barrer, D.M. MacLeod, Faraday Soc. 51 (1955) 1290–1300. [12] A. Lucas et al., Ind. Eng. Chem. Res. 31 (1992) 2134–2140. [13] D.-C. Lin et al., Microporous Mesoporous Mater. 70 (2004) 63–70. [14] H. Youssef, D. Ibrahim, S. Komarneni, Microporous Mesoporous Mater. 115 (2009) 527–534. [15] S. Chandrasekhar, P.N. Pramada, Porous Mater. 6 (1999) 283–297. [16] J. Aguado et al., Microporous Mesoporous Mater. 75 (2004) 41–49. [17] D. Boukadir, N. Bettahar, Z. Derriche, Ann. Chim. Sci. Mater. 227 (2002) 1–13. [18] C.S. Cundy, P.A. Cox, Microporous Mesoporous Mater. 82 (2005) 1–78. [19] Y. Hu, X. Liu, Miner. Eng. 16 (2003) 1279–1284. [20] C. Bebon et al., Microporous Mesoporous Mater. 53 (2002) 13–20. [21] S. Chandrasekhar, P.N. Pramada, Microporous Mesoporous Mater. 108 (2008) 152–161. [22] H.M. Rietveld, Acta Crystallogr. 22 (1967) 151–152. [23] R.A. Young, J. Appl. Crystallogr. 28 (1995) 366–367. [24] L.V. Azároff, Elements of X-Ray Crystallography, first ed., McGraw-Hill, New York, 1968. [25] M.J. CagliotI, A. Paoletti, F.P. Ricci, Nucl. Instrum. Methods 35 (1958) 223–228. [26] S. Chandrasekhar, P.N. Pramada, Ceram. Int. 27 (2001) 351–361. [27] L.I.D. da Silva et al., Quím. Nova 29 (2006) 46–51. [28] S.A. Khan, R.L. Mulvaney, R.G. Hoeft, Soc. Am. J. 64 (2000) 1083–1089. [29] R. Dimitrijevic, W. Lutz, A. Ritzmann, J. Phys. Chem. 67 (2006) 1741–1748. [30] V. Gramlich, Z. Meier, Kristallogr., Kristallgeom., Kristallphys., Kristallchem. 133 (1971) 134–149. [31] C. Breen et al., Vib. Spectrosc. 43 (2007) 366–379. [32] D. Akolkar, A. Chaffee, R.F. Howe, Zeolites 19 (1997) 359–365. [33] R.W. Stevens, R.V. Siriwardane, J. Logan, Energy Fuels 22 (2008) 3070–3079. [34] H. Yamada et al., Sci. Technol. Adv. Mater. 6 (2005) 394–398. [35] Y. Huang, Z. Jiang, Microporous Mater. 12 (1997) 341–345. [36] K.A. Iyer, S.J. Singer, Phys. Chem. 98 (1994) 12679–12686. [38] A.A. Ismail et al., Colloids Surf. A 366 (2010) 80–87. [39] S. Cinar, B. Beler-Baykal, Water Sci. Technol. 51 (2005) 71–77.